SUPERNOVAE AND NEUTRON STARS STIRLING A.COLGATE New Mexico Institute of Mining and Technology, Socorro, N.M. 87801, U.S.A. Abstract. The formation of a neutron star and the event observed as a supernova are presumed by most astrophysicists to be simultaneous. The theoretical difficulty of understanding this is compounded by a hierarchy of conflicting phenomena dominated by the possibility of effective neutrino transport of the heat of the neutron star formation to the outer envelope. A thermonuclear detonation of a carbon core may take place at just the right density such that both neutron star formation and neutrino assisted envelope ejection can occur together. 1. Supernova and Neutron Star: The Same Event? This paper reviews one particular problem of supernova theory; namely, how can the birth of a neutron star and a supernova be one and the same event and yet how can they not be? The theories of supernovae are faced with a major dichotomy: on the one hand, we see a neutron star remnant at the site of one supernova, the Crab, and infer that many other cases are similar - yet on the other hand, serious doubt has been cast on the numerical theoretical calculations that might explain such events. The problem centers on the requirement of simultaneously ejecting something like a solar mass at the same time that a similar mass collapses to a neutron star. First of all, what is the justification for both 'a solar mass' and 'near simultaneity'? The collapse to a neutron star requires that sufficient mass be assembled at a low enough temperature such that the available specific gravitational energy is greater than the specific energy required to convert a proton-electron pair into a neutron, when that proton is at the nuclear binding corresponding to that in helium. At the stellar densities where collapse to a neutron star is imminent, there is little question that stellar nucleosynthesis has proceeded to the point of at least helium - and more likely carbon-oxygen and even to the end point - iron. Ninety percent of the nuclear binding occurs in helium and so whatever additional synthesis has occurred will not signifi­ cantly change the approximate result. The minimum energy required to convert He to free neutrons is about 1019 erg g"1. This must be equated to MG/r, the gravitational potential. If we want to minimize M, we must minimize the radius. The minimum radius in turn is set by the condition that we do not want the degenerate electron pressure to require more work in collapse than the conversion of protons to neutrons - or put another way, protons will be converted to neutrons whenever the electron Fermi level becomes greater than the nuclear energy difference. This occurs at Q~2 x 1011 g cm"3 (^Fermi —10 MeV) which is therefore the density of incipient 7 collapse. Using this density results in r~10 cm and M~MQ. If the temperature is raised (compared with EF~ 10 MeV) the minimum mass will be increased. The result is comfortably assuring in that we know that it requires a mass greater than the Downloaded Cfrom. /. Hansenhttps://www.cambridge.org/core (ed.), Physics of Dense Matter,. IP address: 183-187 170.106.33.22. All rights, onreserved. 02 Oct 2021 at 08:52:00, subject to the Cambridge Core terms of use, Copyright © 1974 by the I A U. available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0074180900100014 184 S. A. COLGATE Chandrasekhar limit 1.13 MQ to overcome electron degeneracy pressure and reach this density in the first place, which, anyhow, is a similar way of looking at the condition for collapse. The estimate of the ejection of a solar mass is an observational estimate derived from the light curves. One observes a peak luminosity and temperature; granted that these observations have some considerable error and even variation as to class of supernova, but for a typical type I supernova the peak luminosity may be 2 x 1043 erg s"1 for a 4 4 surface temperature ^10 K. The surface area required, A(c/4)aT = L9 becomes 2 x 1031 cm2. The absolute minimum mass would be this area divided by the optical opacity. At the necessarily low density, this is Compton opacity, 1032 g for the surface layer. Models describing the mechanism of producing the optical radiation require some finite depth of the expanding (observed by Doppler shift) matter which at a minimum may be 10 mean free paths thick (in order that the diffusive release time of the radiation be as long as the observed width of one week) implying that at least 1 /2 Me is ejected. Finally, we are presuming that these events occur nearly simultaneously. Extra­ polation of the slowing down of the Crab pulsar leads to an origin within years of the supernova event. One could assume ejection took place (say 10 years?) before or after collapse; namely, that the events are only secondarily related, but the theoretical difficulties become even more staggering. If ejection comes after collapse, then the matter must be lifted from the neutron star surface at the bottom of a gravitational well of greater than 100 MeV per nucleon yet be observed 'dribbling' out into space at 1 to 2 MeV per nucleon (maximum bulk expansion velocity ~2 x 109 cm s"1). No energy source has been proposed that could possibly achieve this. Ejection before collapse is more difficult to exclude. Conceivably, a thermonuclear detonation of a carbon-oxygen layer could take, place at the surface of a more highly evolved, say, silicon core. The outer layer might be ejected and the smaller energy release in the core might be too small to disrupt the core and leave a large enough mass that later evolution would lead to collapse. The subtlety of conditions that would lead to thermonuclear detonation and partial disruption with­ out simultaneously triggering collapse would seem hard to arrange - almost contrived - and so unless there is overwhelming evidence to the contrary, most would agree that despite current difficulties, collapse and ejection must occur in the same dynamical event, but it is difficult to manage both at the current sophistication of theoretical analysis. The original concept of achieving both events simultaneously depends upon utilizing the binding energy of the neutron star (^100 MeV per nucleon) to eject an outer, less strongly bound (1 MeV per nucleon) mass fraction. We emphasize net binding energy, the difference between the internal and gravitational energies. If comparable masses are ejected and collapsed, it would seem that the large energy difference would be more than adequate to achieve partial ejection. However, the difficulty arises in the inefficiency of energy transfer. If initially cold matter of relatively small binding energy were to collapse to a neutron star, the difference in binding Downloaded from https://www.cambridge.org/core. IP address: 170.106.33.22, on 02 Oct 2021 at 08:52:00, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0074180900100014 SUPERNOVAE AND NEUTRON STARS 185 energy would have to appear in internal degrees of freedom - oscillations and ultimately heat. Because of the propensity for the 'cold' fluid to form a shock wave during col­ lapse, the formation process is relatively inelastic; the 'bounce' is weak, and essentially all the binding energy appears as heat. This heat, in turn, can be conducted to the outer layer by 'radiation' transport, heating the outer layers to the point where they blow off. The problem is that ordinary Planck radiation diffuses so slowly (the char­ acteristic time is the Helrriholtz contraction time of 105 yr) that the outer layers would have long since collapsed in seconds onto the neutron star core and possibly exceeding the stable mass limit and forming a black hole. Fortunately, for the theoretician, at the high temperature of the just-formed neutron star of ^30 MeV (3 x 1011 K or 100 MeV per nucleon) radiation must include neutrinos as leptons in thermodynamic equilibrium and this equilibrium occurs in times of 10"8 to 10~6 s depending upon density, temperature and which kind of neutrino is involved, electron or muon. The thickness of the imploding matter onto the forming neutron star is variously one to ten mean free paths for the electron neutrinos and much less than one mean free path for the muon neutrinos. Therefore, only the electron neutrinos can transport heat to the outer layers and two conditions can prevent or greatly reduce this heat transport, and hence prevent sizeable mass ejection. If the imploding matter is thick enough (about 10 mean free paths) so that the neutrino radiation diffusion velocity (~c/10) is less than the implosion velocity, then the diffusing heat will be continuously carried back to the neutron star and the outer layers will not be heated. Similarly, if the muon emission is great enough, it will allow the escape of the heat without interaction with the outer layers. Fortunately, one effect argues strongly in favor of some partial heat transport. As the temperature falls, so also does the neutrino electron scattering cross section, 2 4 S (7vocE , and with nuclear thresholds and degeneracy CTVCCE ~ SO that decreasing temperature allows a more rapid (longer mean free path) diffusion of heat. In addition, a lower temperature favors electron neutrino emission as opposed to muon neutrino emission. The actual temperature and ratio of electron to muon neutrinos depend upon details of the hydrodynamics, but some of these details become crucial to the final outcome.